full text

Transcription

full text
Journal of Geosciences, 55 (2010), 57–67
DOI: 10.3190/jgeosci.059
Original paper
Aquamarine beryl from Zealand Station, Canada: a mineralogical
and stable isotope study
Kristy-Lee BEAL*, David R. LENTZ
Department of Geology, University of New Brunswick, Fredericton, New Brunswick, E3B 5A3, Canada; [email protected], [email protected]
*
Corresponding author
Aquamarine crystals occur in crustal A-type pegmatite–aplite dykes and associated quartz veins and greisen zones at
the Zealand Station Be–Mo deposit, south-central New Brunswick (Canada). Electron-probe microanalysis (EPMA)
of aquamarine determined that the chromophore Fe (< 1.4 wt. % FeOT) is present in octahedral sites, with substitutions
responsible for simple and oscillatory zonation (SEM–BSE imaging). Average water content (1.53 wt. %) was calculated using an empirical formula, consistent with two TREL (>800 °C) of 1.3–1.4 wt. % in beryl channels. The δ18O of the
quartz and beryl were within 1 ‰, consistent with magmatic crystallization from an 18O enriched source, and δD values
on channel water have magmatic signatures.
Keywords: aquamarine, pegmatite, aplite, Allandale Granite, Zealand Station, New Brunswick
Received: 16 November 2009; accepted 12 March 2010; handling editor: M. Novák
1.Introduction
Beryllium is a relatively rare element in the Earth’s crust,
ranking 47th most abundant. It averages approximately 3
ppm in the upper crust, which is fairly elevated compared
to 60 ppb inferred in the primitive mantle (Grew 2002).
Beryl, the most common beryllium-bearing mineral in the
Earth’s crust, occurs in granites, granite pegmatites, and
in veins. Beryl can form at any stage of pegmatite consolidation/crystallization, by autometasomatic processes,
and during hydrothermal events (Černý 2002; Wang
et al. 2009). Beryl is a tectosilicate (Aurisicchio et al.
1988), with the ideal formula expressed as Be3Al2Si6O18,
although elemental substitutions in the structure cause
the formula to become more complex (Černý 2002).
Beryl has a hexagonal structure where the six-member
rings of Si-bearing tetrahedra (SiO4) are stacked in the
c-direction and linked laterally with a relative rotation
of 30o between adjacent rings. These rings are linked
in the direction of the a-axis by Be-bearing tetrahedra
(BeO4) and together they stack in the c-axis. Al-bearing
octahedra (O) provide further linkages parallel and
perpendicular to the c-axis (Hawthorne and Humincki
2002; Černý et al. 2003). With all the substitutions considered within the beryl structure, the formula becomes
C
(Na,Cs)2X–Y+Z C(H2O,He,Ar)≤2–(2X–Y+Z)+Na T(2)(Be3–XLiYX–Y)
O
(Al,Fe,Sc,Cr,V)2–Z O(Fe,Mg,Mn)2+Z T(1)Si6O18, where Y ≤
2, X ≥ Y, Z « 2 and 2X–Y + Z ≤ 2 (Černý 2002; Diego
Gatta et al. 2006); C are the channel sites. The substitutions cause distortion in the beryl structure, as well as
variation in the bond-length and bond-angles (Aurisicchio
et al. 1988).
There are several aquamarine deposits in Canada; the
most notable among them is the True Blue aquamarine
showing in the Yukon Territory (Turner et al. 2007). There
are currently fourteen known beryl occurrences in New
Brunswick that are hosted in a number of different environments related to differentiated felsic magmas (Fig. 1).
The Zealand Station Be–Mo deposit, initially staked in
the early 1950s because of a molybdenite occurrence, is
located approximately 25 km northwest of Fredericton.
The aquamarine crystals at the Zealand Station deposit
attain up to 8 cm in diameter and are of clear to milky,
light to medium blue colour and commonly euhedral to
subhedral. The compositions of the beryl grains were
studied to classify the origin of the pegmatite and the
nature of the related hydrothermal fluids.
2.Geological setting
The Zealand Station aquamarine deposit is located along
the northeastern cusp of the Early Devonian Hawkshaw
Granite of the Pokiok Batholith. The Hawkshaw Granite intruded the metasedimentary rocks of the Silurian
Kingsclear Group (Caron 1996). The Pokiok Batholith
consists of five units, from oldest to youngest: Hartfield
Tonalite dated at 415 ± 1 Ma (U-Pb titanite), Hawkshaw
Granite (411 ± 1 Ma, U-Pb titanite), Skiff Lake Granite
(409 ± 2 Ma, U-Pb zircon), Lake George Granodiorite
Stock (412 +5/–4 Ma, U-Pb zircon), and Allandale Granite, which was dated at 402 ± 1 Ma by U-Pb monazite
(Whalen 1993; Whalen et al. 1996). The Pokiok Batholith is the result of the Acadian Orogeny, which is noted
www.jgeosci.org
Kristy-Lee Beal, David R. Lentz
Pabineau
Pabineau
Lake
Lake
CANADA
Pabineau
Pabineau
Lake SE
Lake
SE
NEW BRUNSWICK
Clearwater
Clearwater
Brook
West
Brook West
Beaver
Beaver Brook
Brook
Lake
Lake
South Burnthill
South
Burnthill
Lower
Lower Burnthill
Burnthill
Brook
Brook
Brewers
Brewers
Mills
Mills
Peaked
Peaked
Mountain
Mountain
Trout
Trout Lake
Lake
Pluton
Pluton
Burnthill
Burnthill
South
South Ridge
Ridge
Zealand
Zealand
Station
Station
Fredericton
Misers
Misers Hill
Hill
Deep
Deep Lake
Lake
N
100 km
for associated syn- to post-tectonic granitoid intrusions
from 410 to 380 Ma (Rast and Skehan 1993). This area
of New Brunswick is situated in the Central Mobile Belt
of the Canadian Appalachian Orogen and is viewed as a
result of evolving geological environments involving the
opening and closure of the Precambrian–Early Paleozoic
Iapetus Ocean (Whalen et al. 1996).
The Zealand Station aquamarine deposit is hosted
within an aquamarine-rich section of a SE-trending pegmatite–aplite dyke and in association with hydrothermal
features including quartz veins and greisen zones in the
host monzogranite, inferred to be the Hawkshaw Granite.
This fine- to coarse-grained monzogranite is subdivided
into three different facies based on field observations by
Chrzanowski and Elliott (1986): (1) pink granite, (2) white
granite, and (3) biotite-rich granite (Fig. 2). The petrological and geochemical differences between the individual
intrusive phases are insignificant and all exhibit gradational contacts over 0.5 m to 1 m. All sub-types of the host
monzogranite contain quartz, feldspars, biotite, rutile, and
muscovite with trace zircon, monazite, and xenotime.
The SE-trending pegmatite–aplite dykes crosscut the
host monzogranite and contain an aquamarine-bearing
section in the largest dyke that is approximately 50 m
by 2 m (Fig. 2). The dykes are locally porphyritic with
euhedral quartz (up to 4 cm) and orthoclase (up to 15 cm)
58
Fig. 1 Geologic map of New Brunswick
showing the locations of all known beryl occurrences (DNR-NB bedrock map
NR-1 2005, see http://www.gnb.ca/0078/
m i n e r a l s / P D F / B e d ro c k _ G e o l o g y _
MapNR1-e.pdf for map unit descriptions).
phenocrysts and contain additional quartz, orthoclase,
albite, primary muscovite, trace to local 5 % biotite and
accessory monazite, zircon, apatite, wolframite, and xenotime in the aplitic groundmass. This particular dyke has
been dated at 400.5 ± 1.2 Ma using U-Pb TIMS on magmatic zircon linking this phase to the Allandale Granite of
the Pokiok Batholith (Beal et al. 2010). The aquamarinebearing section (Fig. 3a) contains approximately 30 to
40 % of blue aquamarine crystals (<2 cm wide), albite,
orthoclase, primary muscovite, and minor quartz with
Beal, calcite,
Fig. 1 monazite, zircon, apatite, and xeaccessory rutile,
notime (Fig. 3b). The contacts of the aquamarine-bearing
section with the dyke are sharp and marked locally by
abundant biotite. This aquamarine-bearing section was
dated at 404 ± 8 Ma (U-Th-Pb monazite; Beal et al. 2010)
linking the mineralization to the dyke.
There is a 2 m by 2 m pegmatite outcrop located approximately 50 m east of the mapped area and consists
of a border zone, intermediate zone, and a core of predominately plagioclase and quartz up to 15 cm, biotite
booklets up to 3 cm wide, and wolframite crystals up to
3 cm long. The border and intermediate zones contain
orthoclase, albite, quartz, and primary muscovite with
trace amounts of biotite, zircon, rutile, monazite, chlorite,
calcite, and iron-oxides. No beryl has been identified in
this pegmatitic phase.
Aquamarine from Zealand Station, Canada
46o03'24''.685
66o58'01''.830
06-KB-108
06-KB-107
06-KB-02c
06-KB-106
06-KB-07b
06-KB-105
06-KB-103
06-KB-103b
DEVONIAN
White granite: fine- to coarse-grained granite with white
feldspars and minor biotite
Pink granite: fine- to coarse-grained granite with
microphenocrysts of pink K-feldspar and minor micas
Biotite-granite: fine- to coarse-grained granite with
microphenocrysts of white K-feldspar and up to
20 % vol. biotite
Quartz-alkali feldspar pegmatite-aplite dyke with local
aquamarine-bearing section
Muscovite-quartz greisen: predominantly muscovite and
quartz with yellow to pinkish surface alteration
SILURIAN
Hornfels: metamorphosed light to medium grey shale
with relict foliation. Xenoliths of hornfels are present in
the granite
Sharp contact
Gradual contact
2007 Zealand Station
(aquamarine) deposit:
geology and sample location map
Beryl occurrence
Sample site – ‘vein’ beryl analysis
Sample site –‘dyke’ beryl analysis
Fig. 2 Geology of the Zealand Station aquamarine occurrence (modified
from Chrzanowski and Elliott 1986).
Aquamarine and molybdenite mineralization, with
trace amounts of scheelite, occur locally around greisen
pockets (Fig. 4) that are marked by earthy hematization
on the weathered surface and moderate alteration in the
surrounding host monzogranite. Muscovite grains are
less than 1 mm in size and occur in clusters, whereas
the quartz is very fine grained to cryptocrystalline and
appears laminated. Aquamarine crystals up to 5 cm long
and molybdenite booklets up to 1cm wide occur around
some greisen zones.
Quartz veins contain varied amounts of quartz, feldspar, biotite, muscovite, blue aquamarine (up to 8 cm
wide; Fig. 4), and molybdenite (up to 2 cm wide booklets). The veins range between 0.5 cm to rarely 10 cm
wide and have two predominant orientations: 135/90o and
010/75o W (Chrzanowski and Elliott 1986).
Fig. 3a – Mineralized portion of the pegmatite–aplite dyke showing
06-KB-07A. b – Photomicrograph in crossed polarized light depicting
the relationship between magmatic beryl and surrounding minerals
(06-KB-07A).
Road
a
5cm
b
Beal, Fig. 2
Brl
Ms
Kfs
Brl
Pl
59
Kristy-Lee Beal, David R. Lentz
3.Analytical methods
Fig. 4 Euhedral Zealand Station beryl crystals collected by a local prospector from one of the greisen veins (left) and a quartz vein (right).
The JEOL-733 Electron Probe Microanalyser (EPMA)
at the University of New Brunswick’s Microscopy and
Microanalysis Facility was used for the examination of
aquamarine in nine large polished thin sections. The
microprobe could not detect volatiles or elements lighter
than boron. Operating conditions were set to a beam
current of 30 nA and the data for samples were collected
for 200 s (for Cs, Fe (total), Mn, Cr, V, Sc, and F), 100 s
(Ca, Mg, Na, K), and 30 s (Al and Si). The following
standards were used: fluorite (F), jadeite (Na), olivine
(Mg), bytownite plagioclase (Al, Si, Ca), orthoclase (K),
vanadium metal (V), scandium metal (Sc), chromite (Cr),
bustamite (Mn), hornblende (Fe), and pollucite (Cs).
The analytical results, acquired by the EPMA, were
recalculated from weight percent oxide to atoms per
Tab. 1 Average EPMA data of the zoned aquamarines from the Zealand Station
hydrothermal dyke hydrothermal dyke hydrothermal dyke
core
core
middle
middle
rim
rim
average
average
average
average
average
average
n
11
4
6
2
11
4
SiO2 (wt.%)
67.65
67.49
67.00
67.66
67.71
67.84
beryl
scan
#1
1
67.28
beryl
scan
#6
1
66.45
beryl
scan
#12
1
66.63
beryl
scan
#18
1
66.87
beryl
scan
#24
1
66.44
beryl
scan
#31
1
67.04
beryl
scan
average
31
66.43
Al2O3
17.48
16.98
17.11
16.93
17.39
17.55
17.54
17.04
17.32
17.51
17.38
17.78
17.30
0.03
sample*
Sc2O3
0.01
0.04
0.03
0.06
0.03
0.03
0.01
0.04
0.03
0.03
0.02
0.01
V 2O 3
0.07
0.24
0.04
0.17
0.05
0.23
0.09
0.10
0.10
0.05
0.09
0.07
0.08
Cr2O3
BeO*
MgO
CaO
FeOT
Na2O
0.01
13.62
0.42
0.01
0.83
0.33
0.00
13.59
0.54
0.00
1.16
0.46
0.01
13.40
0.39
0.01
1.00
0.39
0.00
13.52
0.51
0.01
1.22
0.40
0.01
13.54
0.39
0.01
0.83
0.32
0.00
13.72
0.38
0.01
0.77
0.34
0.00
11.50
0.49
0.00
0.83
0.34
0.00
11.35
0.47
0.00
1.15
0.32
0.01
11.39
0.44
0.00
0.97
0.31
0.00
11.43
0.32
0.00
0.92
0.25
0.00
11.37
0.39
0.01
1.02
0.28
0.02
11.43
0.24
0.00
0.27
0.17
0.01
13.77
0.40
0.00
0.96
0.30
0.06
0.06
0.05
0.06
0.05
0.02
0.04
0.07
0.05
0.05
0.05
0.01
0.05
0.03
0.01
100.54
1.54
0.09
0.00
100.65
1.65
0.07
0.01
99.49
1.59
0.10
0.00
100.64
1.60
0.08
0.00
100.41
1.53
0.06
0.01
100.95
1.55
0.03
0.01
98.15
1.55
0.10
0.01
97.10
1.53
0.10
0.00
97.34
1.52
0.07
0.01
97.52
1.47
0.09
0.00
97.13
1.50
0.04
0.00
97.08
1.40
0.08
0.01
99.42
1.51
K 2O
Cs2O
MnO
Total
H2O**
Si4+
(apfu)***
Al3+
Sc3+
V3+
Cr3+
Mg2+
Ca2+
Fe3+
Na+
K+
Cs+
Mn2+
Be2+
6.040
6.040
6.047
6.050
6.048
6.035
6.026
6.032
6.026
6.029
6.021
6.043
6.027
1.840
0.001
0.005
0.001
0.056
0.001
0.048
0.058
0.007
0.001
0.000
3.000
1.790
0.004
0.017
0.000
0.072
0.001
0.068
0.080
0.007
0.004
0.000
3.000
1.821
0.002
0.003
0.000
0.052
0.001
0.052
0.067
0.006
0.003
0.001
3.000
1.784
0.005
0.012
0.000
0.068
0.001
0.079
0.070
0.007
0.004
0.001
3.000
1.831
0.003
0.003
0.001
0.052
0.000
0.049
0.055
0.006
0.003
0.000
3.000
1.840
0.003
0.017
0.000
0.051
0.001
0.045
0.058
0.003
0.002
0.001
3.000
1.852
0.001
0.006
0.000
0.065
0.000
0.062
0.060
0.005
0.001
0.000
3.000
1.823
0.003
0.007
0.000
0.064
0.000
0.087
0.056
0.008
0.004
0.000
3.000
1.846
0.002
0.007
0.001
0.059
0.000
0.073
0.054
0.006
0.004
0.000
3.000
1.861
0.002
0.004
0.000
0.044
0.000
0.070
0.044
0.006
0.003
0.001
3.000
1.857
0.002
0.007
0.000
0.052
0.001
0.077
0.050
0.005
0.003
0.000
3.000
1.889
0.001
0.005
0.001
0.032
0.000
0.021
0.029
0.001
0.001
0.000
3.000
1.850
0.003
0.006
0.001
0.053
0.000
0.072
0.051
0.006
0.003
0.000
3.000
Notes: *‘dyke beryl’ refers to beryl in the aquamarine-bearing section of the aplite dyke and ‘hydrothermal beryl’ to the beryl hosted in quartz
veins and near greisenized zones.**H2O is estimated from H2O wt. % = (Na2O in wt. % + 1.4829)/1.1771 (Guiliani et al. 1997). ***Compositions
were recalculated on the basis of 3 Be and 18 O atoms per formula unit (apfu).
60
Aquamarine from Zealand Station, Canada
4.Mineralogy
The aquamarine crystals at the Zealand Station deposit
are clear to milky, light to medium blue in colour, commonly euhedral to subhedral, and up to 8 cm wide. There
is rare zoning from dark blue in the core of the crystal
to light blue in the rim seen in the aquamarine-bearing
section of the dyke. The backscattered electron (BSE)
images and electron microprobe study revealed simple
compositional zonation in the beryl grains with euhedral
to anhedral cores from both the dyke and hydrothermal
sources. A total of 38 electron microprobe analyses were
obtained from 12 individual beryl crystals and divided
into ‘dyke beryl’ and ‘hydrothermal beryl’, the latter
encompassing beryl found in both the quartz veins and
greisen zones. Additional 32 analyses were obtained
from one beryl crystal from a quartz vein (06-KB-108)
that displayed oscillatory zonation (Table 1). The EPMA
results were recalculated based on 18 O and 3 Be apfu
giving the maximum amount of Be content and ignoring
other possible substitutions in the Be-site.
Changes in the colour of beryl crystals correlate with
changes in the composition (Hammarstrom 1989). The
chromophores for many beryl gemstones include Cr ±
V in green emerald, Mn in pink morganite, Fe3+ in yellow heliodor, and Fe2+ in blue aquamarine (Vianna et al.
2002a, b; Mihalynuk and Lett 2003), although Figueiredo
et al. (2008) noted variable Fe 3+/Fe 2+ in aquamarine.
The measured Cr and V average concentrations were
0.0006 and 0.007 apfu, respectively, so the dominant
chromophore is Fe. Thus the Zealand Station beryl can
be classified as aquamarine.
The primary occupant of the octahedral site, Al3+, contains a maximum of 2 atoms, displays a wide variation
(1.745 to 1.910 apfu). The deficiency in the octahedral
site is related to the mutual substitution of Fe2+, 3+, Mg2+,
Mn2+, Sc3+, V3+, and Cr3+. The expected negative trend
Mg+Sc+V+Cr+Mn+Fe
(apfu)
a)
0.3
0.25
0.2
0.15
0.1
0.05
0
1.7
1.75
1.8
1.85
1.9
1.95
2
1.9
1.95
2
Al (apfu)
b)
Mg+Sc+V+Cr+Mn+Fe+
Excess Si (apfu)
formula unit (apfu) with the aid of software called ‘Formula’ written by Dr. T.S. Ercit of the Canadian Museum
of Nature (1996).
Hand-picked separates of quartz and beryl from three
dyke, vein, and greisen zone samples were analyzed for
oxygen isotopes; hydrogen isotopes were analyzed from
the channel H2O of beryl. Detailed analytical methods
were described by Kyser et al. (1999). Estimated errors
on reported ‰ values relative to V–SMOW are < 0.2
‰ for oxygen and < 3 ‰ for hydrogen isotopes. The
samples were heated to temperatures greater than 800 °C
to release the channel H2O (see Brown and Mills 1986;
Fallick and Barros 1987); isotopic fractionation during
the dehydration is noted to be minimal (Taylor et al.
1992).
0.3
0.25
0.2
0.15
0.1
0.05
0
1.7
1.75
1.8
1.85
Al (apfu)
Fig. 5a – Binary plot of Al (apfu) content versus the sum of other cations in Zealand Station beryl that would substitute into the octahedral
site. b – Al (apfu) content versus the sum of other cations plus excess
Si. Symbols are scan (black circle), dyke core (black triangle), dyke
middle (black square), dyke rim (black rim), hydrothermal core (grey
triangle), hydrothermal rim (grey square), and hydrothermal rim (grey
triangle).
from the Zealand aquamarine samples is offset below the
1:1 ratio trend by approximately 0.04 apfu (Fig. 5a). The
Si4+ analyses have a maximum of 6.061 apfu in a tetrahedral site that permits only 6 atoms. If the excess Si is fit
into the Al-bearing octahedral site as another substituting
cation (Fig. 5b), the Al site is filled along the expected
trend and the data show less variance.
The valency of the iron can be determined by balancing the monovalent cations (Na+, K+, 2Ca2+, and Cs+) of
the beryl channels with the divalent cations (Mg2+, Mn2+,
and Fe2+) of the octahedral site to a 1:1 trend. It should
be noted that Ca2+ is included within the monovalent
cations, because alkalis in the channel-sites should equal
the divalent cations that substitute for Al3+, but there is
a charge balance deficit in the Al3+-bearing site. The
EPMA data from the Zealand Station aquamarine show
variable right lateral offset (between 0.00 to 0.075 apfu)
to the expected trend (Fig. 6a). In Fig. 6b the Fe has been
removed from the divalent cations assuming the iron is
in the Fe3+ form. The deviation is possibly due to small
amounts of Fe2+ within the channels, but could also be
due to overestimation of the monovalent cations. This assumption was also completed for aquamarine from True
Blue, Yukon (Turner et al. 2007).
Alkalis, Cs, He, Ar, and water have the ability to be
positioned within the beryl channels. It can be assumed
61
Kristy-Lee Beal, David R. Lentz
2Ca+K+Cs+Na (apfu)
a)
a)
Scan #32
0.14
0.12
0.1
0.08
0.06
0.04
0.02
0
0
0.05
0.1
0.15
0.2
0.25
Scan #01
Mn+Fe(T)+Mg (apfu)
0.14
0.12
0.1
0.08
0.06
0.04
0.02
0
1 mm
b)
0
0.02
0.04
0.06
0.08
0.1
Mn+Mg (apfu)
Fig. 6a – Binary plot of monovalent cations versus divalent cations in
Zealand Station beryl. b – Monovalent cations versus divalent cations
variations; all of the Fe in the Al site is assumed to be Fe3+. Symbols
are the same as Fig. 5.
Al (apfu)
2Ca+K+Cs+Na (apfu)
b)
1.900
1.890
1.880
1.870
1.860
1.850
1.840
1.830
1.820
1.810
1.800
0
5
10
15
20
25
30
35
Position
that the beryl from Zealand contains water either in the
form of H2O and/or OH. According to the empirical relationship of Giuliani et al. (1997; as defined in Brand et
al. 2009), the H2O content can be obtained from:
H2O wt. % = (Na2O wt. % + 1.4829)/1.1771.
Thus the average water content in the Zealand Station
aquamarine samples could be calculated as 1.52 wt. %.
The hydrothermal beryl crystal that displayed oscillatory zonation is 5.5 mm in diameter and is euhedral
on two sides, whereas the other sides are anhedral and
in contact with sericitized plagioclase, muscovite, and
quartz (Fig. 7a). The individual zones, approximately 0.1
to 0.75 mm wide, show short-range compositional variation. Many of the elements are below detection limit, but
FeT, Mg2+, Na+, Si4+, and Al3+ show clear variation in their
concentrations. Many of these elements are associated
with substitutions within the Al-bearing octahedral site
(Fig. 7b). The Al content shows a general negative correlation to other elements that substitute for it, as well as
Na+ that would provide charge balance in the channels.
5.Stable isotope data
In most samples, δ 18O in quartz is greater than other
rock-forming minerals that are more easily altered (cf.
62
Fig. 7a – Back-scattered electron image of an aquamarine 06-KB-108 in
a greisenized zone showing oscillatory zonation and the path for analyses. b – Al variation along the profile that is related to the substitution
present in the octahedral site.
Taylor and Sheppard 1986), so only quartz and beryl were
included in this study. As noted by Taylor et al. (1992),
the H2O released at temperatures in the excess of 800 °C
(TREL) is consistent with the retention of water in beryl.
The molecular form of water in beryl (channel water)
allows its use as a proxy for magmatic fluid isotopic composition in granites and granitic pegmatites (see synopsis
6 for the measurement of the
in Taylor et Beal,
al. 1992),Fig.
mainly
hydrogen isotopic composition.
The δ18O values of quartz are high in the pegmatite–
aplite sample, but within the range for normal granites
(Whalen et al. 1996); these values are comparable to the
δ18O of the pegmatite–aplite beryl–quartz pair analyzed
(∆ 0.2 ‰). These δ18O values are indistinguishable from
the whole-rock values for the nearby Lake George Granodiorite (10.2 ‰), Hawkshaw Granite (9.6 ‰), and
Allandale Granite (9.5 ‰) (Whalen et al. 1996; Yang et
al. 2004). Whalen et al. (1996) also reported δ18O values
for quartz (11.9 ‰), feldspar (9.6 ‰), muscovite (8.2 ‰),
and biotite (5.3 ‰) for the Allandale Granite, which are
consistent with this study; these values, if in equilibrium,
would reflect crystallization or quenching temperatures
Aquamarine from Zealand Station, Canada
to just under 500 °C (cf. Taylor et al. 1979). Such high
δ18O values evidently reflect derivation by contamination
with an 18O-enriched crustal source, typically supracrustal
rocks (Longstaffe 1982; Taylor and Sheppard 1986) or
specifically as a result of O isotope exchange with the
wall rocks prior to final emplacement (see Taylor et al.
1979). The δ18O values for the hydrothermal quartz in
veins and greisens are typically higher. However, with
no T estimates, the equivalent H2O cannot be calculated,
but reflect high-T metasomatism. The beryl–quartz δ18O
pair was analysed (∆ 1.2 ‰) from hydrothermal quartz
(> 600 °C; see equation in Taylor et al. 1992).
The two δD of channel H2O (TREL > 800 °C) in beryl
from the Zealand pegmatite–aplite dykes (H2O yields of
1.3 to 1.4 wt. %) had values from –76 to –67 ‰ (Tab. 2).
The δD of H2O of channel beryl is similar to exsolved
magmatic water or molecular water still dissolved in the
crystallizing melt (i.e., vapour undersaturated, see Taylor
1986). However, these values are generally higher than
emeralds formed at lower temperatures (<< 500 °C)
(Groat et al. 2008), except for those emeralds to green
beryl closely associated with pegmatite-derived fluids
(see Laurs et al. 1996).
6.Discussion
6.1.Petrogenetic aspects
The Zealand Station granitic pegmatite–aplite dykes
are predominately composed of quartz, albite, and Kfeldspar with minor aquamarine, muscovite, wolframite,
and biotite and trace apatite, monazite, and zircon. Beal
et al. (2010) concluded that the pegmatite–aplite dykes,
and associated mineralization, are geochronologically
and geochemically related to the Allandale Granite, the
youngest and most fractionated phase of the Pokiok
Batholith. The pegmatite–aplite dykes are highly fractionated compared to the parental granite, based on high
SW
Beryllium within
magmatic fluids
moving out from
granite with
hydrothermal fluids
Quartz vein
Host granite
Greisen
pocket
Pegmatite
Aplite
Metasediments
Evolved granite
Fig. 8 Proposed genetic model for the aquamarine mineralization
within the Zealand Station deposit. The blue hexagons represent beryl
mineralization (modified after Groat et al. 2007).
incompatible elemental abundances. The dykes show
characteristics typical of crustally-derived A-type granites
contaminated by metasediments of the recently thickened
crust as the magma was emplaced. This resulted in hybrid
characteristics between NYF and LCT families of the
REE pegmatite class for these dykes (Beal et al. 2010).
The heterogeneous nature of the aquamarine gem
pocket in the pegmatite–aplite dyke could be explained
by a late-stage exsolution of a fluid/pegmatitic phase
from volatile-rich magma (Fig. 8). The water pressures
would have been fairly high at their time of formation, so
the aqueous pegmatitic phase coexisted with the fractionated granite (aplite) melt (Jahns and Burnham 1969). The
presence of abundant beryl within one gem zone, rather
than all pegmatitic zones, suggests that the fluid carry-
Tab. 2 Stable isotopic data of selected samples: δD of beryl channel H 2O content and δ18O of beryl and quartz.
Sample
Source
Material
06-KB-07a Beryl
Aplite Dyke
06-KB-103 Beryl
Quartz Vein
Sample
wt. % H2O
δD(VSMOW)
Channel water
1.4
-76
Channel water
1.3
-67
Yield
δ18O(VSMOW)
Material
06-KB-07a
Aplite Dyke
Beryl
15.7
10.6
06-KB-07a
Aplite Dyke
Quartz (<10% contamination)
15.6
10.4
06-KB-103
Quartz Vein
Beryl
16.1
10.2
06-KB-103
Quartz Vein
Quartz (~15% contamination)
15.7
11.4
06-KB-09a
Greisen Zone
Quartz (~15% contamination)
16.1
12.6
63
Kristy-Lee Beal, David R. Lentz
ing the components necessary for beryl formation was
heterogeneous. The gem zone is not as fully developed
as those seen on Elba Island in Italy and Mt. Mica in
Maine, US, suggesting that the separate fluid phase did
not evolve sufficiently to form open miariolitic cavities
before crystallization (cf. Simmons 2007).
The aquamarine present within certain vein sets could
have formed from a supercritical fluid exsolving from the
magma during progressive crystallization (Candela 1997).
The veins follow a joint set that has an approximate
NW–SE trend but deviations exist. Greisen is related to
fluids associated with the dykes in the exocontact zone
of a beryllium-bearing intrusion at depth. Greisen formed
presumably by a reaction of a high-temperature, siliceous
magmatic fluid with elevated F concentrations. This
resulted in autometasomatic alteration within the dykes
and along joints at lower temperatures (~300–550 °C)
crystallizing mica, feldspar, quartz–feldspar, and quartz
veins (cf. Kievlenko 2003; Turner and Groat 2007).
6.2.Beryl systematics
The aquamarine crystals located in the dyke and hydrothermal sources show similar elemental substitutions
and properties; they also indicate decreased vacancysite substitutions in the rim compared to the core. The
decreasing occupancy in the channel site is more notable
in the aquamarine-rich section of the dyke as the channel
substituents contents decrease continuously from 0.091
apfu in the core to 0.064 apfu in the rim; the aquamarine
from hydrothermal sources decrease from 0.066 apfu in
the core to 0.064 apfu in the rim. The oscillatory-zoned
crystal reveals decreased vacancy-site substitution
(average 0.060 apfu) compared to other simply-zoned
aquamarine of the hydrothermal and dyke sources with
a decrease from the core (0.066 apfu) to the rim (0.032
apfu). This type of channel-site zonation is common
in beryl from the Bikita granitic pegmatite, Zimbabwe
(Černý 2002). The aquamarine from the dyke contains
increased H2O (average 1.60 wt. %) compared to the
aquamarine from hydrothermal sources (1.55 wt. %).
The oscillatory-zoned aquamarine contains up to 1.66
wt. % H2O with an average of 1.51 wt. %. According to
Taylor et al. (1992), beryls from granitic pegmatites range
between 1.2 and 2.55 wt. % H2O.
The dyke-hosted aquamarine crystals also show
decreased substitution in the octahedral-site as the Alcontent increases from 1.79 apfu in the core to 1.84 apfu
in the rim. This type of zoning is similar to that described
by Abdalla and Mohamed (1999) for beryl associated
with the Pan-African Belt in Egypt, but is opposite to
that recorded by other authors where the zoning is explained by the exchange reaction with other minerals
present (Franz et al. 1986). Additional role is played by
64
a decrease of temperature and pressure during, or after,
uplift of the dyke (Aurisicchio et al. 1988). Aquamarines
from both the aquamarine-rich section of the dyke and
hydrothermal sources show a greater concentration of Al
in the immediate vicinity of microfractures; this type of
alteration also occurs in the emeralds of Carnaiba, Bahia
State, Brazil (Giuliani 1988).
The presence of excess silica is common when the
Be-site (3 atoms) or Al-site (2 atoms) is assumed full
to calculate the beryl formula (Sanders and Doff 1991;
Abdalla and Mohamed 1999; Groat et al. 2002; Černý
et al. 2003; Turner et al. 2007). Excess silica can be assigned to the Be tetrahedron (Aurisicchio et al. 1988)
permitting the coupled substitution Be + 2Al = Si + 2(Fe,
Mg) (Sanders and Doff 1991). The silica standard used
for EPMA calibration may have caused the excess silica
issue for a number of analyses with excess silica in the
calculations.
Average Fe concentrations in the Zealand Station
aquamarine crystals range from 0.83 to 1.22 wt. % FeOT,
with a maximum iron content of 1.4 wt. % FeOT that is
assumed to be predominately in the Fe3+ state. Cores are
enriched in iron relative to rims of the individual crystals.
The aquamarine in the dykes contains increased average
Fe (0.064 apfu), and V (0.015 apfu) compared to the
aquamarine in hydrothermal sources (0.050 and 0.004
apfu, respectively). The iron content of the Zealand Station aquamarine is generally lower than in other reported
aquamarines; the highest reported value in beryl was from
dark blue aquamarine, True Blue in Yukon, Canada that
contained up to 5.92 wt. % FeOT (Turner et al. 2007).
How chromophores affect the colour of minerals,
particularly beryl, is not well-understood. According to
Mihalynuk and Lett (2003), Fe can occupy many sites
within the structure to produce colour; Fe3+ within the
octahedral site yields a yellow colour, while Fe2+ in the
same site has no effect. It is the presence of Fe2+ within
channels that produces a deep blue colour, characteristic
of aquamarine gemstones (Mihalynuk and Lett 2003).
Others, particularly Goldman et al. (1978) and Rossman
(1981), explained that blue, green, and yellow tones are
the result of different Fe2+/Fe3+ ratios (see Černý 2002).
Another suggestion is that intervalence charge transfer
(IVTC) occurs between Fe2+ and Fe3+, whereby the Fe2+
is located at the Al sites and the Fe3+ is located at an
interstitial octahedral position that lies between the Alsites and is normally empty. It is well documented that
the long-term exposure of beryl to light and associated
radiation will cause colour fading, because of IVTC
(Samoilovich et al. 1971; Nassau 1978; Mathew et al.
2000; Groat et al. 2005).
The aquamarine at the Zealand Station, with an empirically derived average water content of 1.52 wt. %, crystallized between 600 and 660 °C at 2.5 ± 0.5 kbar (200–300
Aquamarine from Zealand Station, Canada
MPa) according to experimental constraints provided by
Pankrath and Langer (2002). This is consistent with the
H2O yield (1.3 and 1.4 wt. %) from the two H isotope
analyses of channel beryl and with a modified liquidus
and solidus relations in extremely fractionated pegmatite
during hypabyssal emplacement (see London 2005).
7.Conclusions
The Zealand Station deposit contains locally abundant
beryl mineralization; there are also variable amounts
of molybdenite, wolframite, and scheelite. The beryl
samples collected all have microfractures, although the
presence of gem-quality beryl samples is possible beneath
the zone of deep weathering. Wolframite and scheelite
mineralization is present locally, although molybdenite
is rather abundant within the magmatically-derived
quartz veins and around greisen pockets in the granitic
host rock.
Other highlighted conclusions from the Zealand Station aquamarine occurrence:
1. The electron-microprobe analyses of Zealand Station
aquamarine in both dykes and veins revealed that the
main chromophore is Fe3+ within the Al-bearing site
with a maximum of 1.4 wt. % FeO T present. This justifies the use of the term ‘aquamarine’ to describe this
beryl. Zonation was visible in all crystals in the BSE
image.
2. The aquamarine crystals contain more silica than the
Si-bearing tetrahedral site permits; the excess silica
fits perfectly as a substituent for Al within in the Albearing octahedral site; however, this is likely a result
of the silica standard used to calibrate the EPMA. It is
recommended to introduce a new beryl standard with
known silica content so the common issue of excess
silica could be properly addressed.
3. Additional geochemical features of the aquamarine
crystals include increased octahedral-site substitutions
from the core to the rim in the magmatic aquamarine and
increased channel-site substitutions from the core to rim
in both the magmatic and hydrothermal aquamarine.
4. The average water content was calculated based on the
equation H2O wt. % = (Na2O wt. % + 1.4829)/1.1771
at 1.53 wt. %, similar to the yields of 1.3 and 1.4
wt. % for the channel water.
5. The high δ18O values for quartz and beryl are consistent with a considerable crustal component to the
magmas and the lower δD values of the channel water
in beryl reflect crystallization from a volatile-undersaturated magmatic source before dyke emplacement.
Acknowledgements The valuable suggestions and constructive comments on a previous version of this manu
script by Drs. Steve R. McCutcheon and Kathleen Thorne
are appreciated. We are very grateful to Prof. Lee Groat
(UBC) for his guidance in beryl analyses and structure
interpretations, and Dr. Doug Hall helped with EPMA and
SEM imaging at UNB. A special thank you is extended
to L. Lawrence for giving me access to the property and
helping with cleaning the outcrops. We thank Dr. Jane
Hammarstrom and Prof. Milan Novák for constructive
reviews and suggestions. Funding for this project was
provided by the Department of Natural Resources – Minerals Division and a Natural Sciences and Engineering
Research Council of Canada – Discovery Grant to David
Lentz.
References
Aurisicchio C, Fioravanti G, Grubessi O, Zanazzi PF (1988)
Reappraisal of the crystal chemistry of beryl. Amer Miner
73: 826–837
Abdalla HM, Mohamed FH (1999) Mineralogical and geochemical investigation of emerald and beryl mineralization, Pan-African Belt of Egypt: genetic and exploration
aspects. J Afr Earth Sci 28: 581–598
Beal K, Lentz DR, Hall D, Dunning G (2010) Mineralogical, geochronological, and geochemical characterization
of the Early Devonian aquamarine-bearing dykes of the
Zealand Station Be–Mo deposit, west-central NB. Can
J Earth Sci 47: in press
Brand AA, Groat LA, Linnen RL, Garland MI, Breaks
FW, Giuliani G (2009) Emerald mineralization associated with the Mavis Lake pegmatite group, near Dryden,
Ontario. Canad Mineral 47: 315–336
Brown Jr. GE, Mills BA (1986) High-temperature structure
and crystal chemistry of hydrous alkali-rich beryl from
the Harding pegmatite, Taos County, New Mexico. Amer
Miner 71: 547–556
C andela PA (1997) A review of shallow, ore-related
granites: textures, volatiles, and ore metals. J Petrol 38:
1619–1633
Caron A (1996) Geology of the Pokiok Batholith aureole,
with emphasis on the Lake George Mine, York County,
New Brunswick. Natural Resources and Energy, Geoscience Report 94-2: 1–91
Chrzanowski MJ, Elliott CG (1986) Progress report beryllium Zealand Station, N.B. (21J/02). New Brunswick
Department of Natural Resources Division, Open File
473350, pp 1–18
Černý P (2002) Mineralogy of beryllium in granitic pegmatites. In: Grew ES (ed) Beryllium – Mineralogy,
Petrology and Geochemistry. Rev Mineral Geochem
50: 405–444
Černý P, Anderson AJ, Tomascak PB, Chapman R (2003)
Geochemical and morphological features of beryl from
65
Kristy-Lee Beal, David R. Lentz
the Bikita granitic pegmatite, Zimbabwe. Canad Mineral
41: 1003–1011
Diego Gatta G, Nestola F, Bromiley GD, Mattauch S
(2006) The real topological configuration of the extraframework content in alkali-poor beryl: a multi-methodological study. Amer Miner 91: 29–34
Fallick AE, Barros JG (1987) A stable-isotope investigation into the origin of beryl and emerald from the Porangatu deposits, Goias State, Brazil. Chem Geol (Isot
Geosci Sect) 66: 293–300
Figueiredo MO, Pereira da Silva T, Veiga JP, Leal Gomes,
C, De Andrade V (2008) The blue colouring of beryls
from Licungo, Mozambique: an X-ray absorption spectroscopy study at the iron K-edge. Mineral Mag 72:
175–178
Franz G, Grundmann G, Ackermand D (1986) Rockforming beryl from a regional metamorphic terrain
(Tauern Window, Austria). Tschermaks mineral petrogr
Mitt 35: 167–192
Giuliani G (1988) Scanning electron microscopy (SEM) and
its applications: determination of solid daughter minerals in fluid inclusions. Congresso Latino-Americano de
Geologia 7: 445–458
Giuliani G, France-Lanord C, Zimmermann JL, Cheil-Letz
A, Arboleda C, Charoy B, Coget P, Fontan F, Giard D
(1997) Fluid composition, δD of channel H2O, and δ18O
of lattice oxygen in beryls: genetic implications for Brazilian, Colombian, and Afghanistani emerald deposits.
Int Geol Rev 39: 400–424
Goldman DS, Rossman GR, Parkin KM (1978) Channel
constituents in beryl. Phys Chem Miner 3: 225–235
Grew ES (2002) Mineralogy, petrology, and geochemistry of
beryllium: an introduction and list of beryllium minerals.
In: Grew ES (ed) Beryllium – Mineralogy, Petrology and
Geochemistry. Rev Mineral Geochem 50: 1–76
G roat LA, M arshall DD, G iuliani G, M urphy DC,
Piercey SJ, Lambor JL, Mortensen JK, Ercit TS, Gault
RA, Mattey DP, Schwarz D, Maluski H, Wise MA,
Wengzynowski W, Eaton DW (2002) Mineralogical and
geochemical study of the Regal Ridge emerald showing,
southeastern Yukon. Canad Mineral 40: 1313–1338
Groat LA, Hart CJR, Lewis LL, Neufeld HLD (2005)
Emerald and aquamarine mineralization in Canada.
Geosci Can 32: 65–76
Groat LA, Giuliani G, Marshall DD, Turner D (2007)
Emerald. In: Groat LA (ed) Geology of Gem Deposits.
Mineralogical Association of Canada 37: 79–100.
Groat LA, Giuliani G, Marshal DD, Turner D (2008)
Emerald deposits and occurrences: a review. Ore Geol
Rev 34: 87–112
Hammarstrom JM (1989) Mineral chemistry of emeralds
and some associated minerals from Pakistan and Afghanistan: an electron microprobe study. In: Kazmi AH, Snee
LW (eds) Emeralds of Pakistan: Geology, Gemology,
66
and Genesis. Geological Survey of Pakistan and Van
Nostrand Reinhold Co. Inc., New York, pp 125–150
H awthorne FC, H uminicki DMC (2002) The crystal
chemistry of beryllium. In: Grew ES (ed) Beryllium –
Mineralogy, Petrology and Geochemistry. Rev Mineral
Geochem 50: 333–404
Jahns RH, Burnham CW (1969) Experimental studies of pegmatite genesis; I, A model for the derivation and crystallization of granitic pegmatites. Econ Geol 64: 843–864
Kievlenko EY (2003) Beryl. In: Soregaroli A (ed) Geology
of Gems. Ocean Pictures Ltd., Colorado, pp 71–119
Kyser TK, O’Hanley DS, Wicks FJ (1999) The origin of
fluids associated with serpentinization processes: evidence from stable-isotope compositions. Canad Mineral
37: 223–237
Laurs BM, Dilles JH, Snee LW (1996) Emerald mineralization and metasomatism of amphibolite, Khaltaro
granitic pegmatite hydrothermal vein system, Haramosh Mountains, northern Pakistan. Canad Mineral 34:
1253–1286
London D (2005) Granitic pegmatites: an assessment of
current concepts and directions for the future. Lithos
80: 281–303
Longstaffe FJ (1982) Stable-isotopes in the study of granitic
pegmatites and related rocks. In: Černý P (ed) Granitic
Pegmatites in Science and Industry. Mineral Association
of Canada, Short Course 8: 373–404
Mathew G, Karanth RV, Gundu Rao RK, Deshpande RS
(2000) Colouration in natural beryls: a spectroscopic
investigation. J Geol Soc India 56: 285–303
Mihalynuk MG, Lett R (2003) Composition of Logtung
beryl (aquamarine) by ICPES/MS: a comparison of beryl
worldwide. In: Geological Fieldwork 2003. Crown Publications Inc., British Columbia, pp 141–146
Nassau K (1978) The origins of color in minerals. Amer
Miner 63: 219–229
Pankrath R, Langer K (2002) Molecular water in beryl,
VI
Al2[Be3Si6O18]·nH2O, as a function of pressure and
temperature: an experimental study. Amer Miner 87:
238–244
Rast DC, Skehan JW (1993) Mid-Paleozoic orogenesis in
the North Atlantic: the Acadian Orogeny. In: Roy DC,
Skehan JW (eds) The Acadian Orogeny: Recent Studies
in New England, Maritime Canada, and the Autochthonous Foreland. Geol Soc Spec Papers 275: 1–26
Rossman GR (1981) Origin of color in pegmatite minerals.
In: Brown GE (ed) The Mineralogy of Pegmatites. Mineralogical Society of America – Friends of Mineralogy
Symposium, Tuscon, Arizona, pp 1–6
Samoilovich MI, Tsinober LI, Dunin-Barkovskii RL (1971)
Nature of the coloring in iron-containing beryl. Sov Phys
– Crystallography 16: 147–150
Sanders IS, Doff DH (1991) A blue sodic beryl from southeast Ireland. Mineral Mag 55: 167–172
Aquamarine from Zealand Station, Canada
Simmons WB (2007) Gem-bearing pegmatites. In: Groat
LA (ed) Geology of Gem Deposits. Mineralogical Association of Canada 37: 169–206
Taylor BE (1986) Magmatic volatiles: isotopic variation of
C, H, and S. In: Valley JW, Taylor HP Jr, O’Neil JR
(eds) Stable Isotopes in High Temperature Geological
Processes. Mineral Soc Am Rev Mineral 16: 185–225
Taylor BE, Foord EE, Friedrichsen HE (1979) Stable
isotope and fluid inclusion studies of gem-bearing granitic pegmatite–aplite dikes, San Diego Co., California.
Contrib Mineral Petrol 68: 187–205
Taylor HP Jr, Sheppard SMF (1986) Igneous rocks: I. Processes of isotopic fractionation and isotopic systematics.
In: Valley JW, Taylor HP Jr, O’Neil JR (eds) Stable
Isotopes in High Temperature Geological Processes.
Mineral Soc Am Rev Mineral 16: 227–271
Taylor RP, Fallick AE, Breaks FW (1992) Volatile evolution in Archean rare element granitic pegmatites: evidence from the hydrogen isotopic composition of channel
H2O in beryl. Canad Mineral 30: 877–893
Turner D, Groat LA (2007) Non-emerald gem beryl. In:
Groat LA (ed) Geology of Gem Deposits. Mineralogical
Association of Canada 37: 111–143
Turner D, Groat LE, Hart CJR, Mortensen JK, Linnen RL,
Giuliani G, Wengzynowski W (2007) Mineralogical and
geochemical study of the True Blue aquamarine showing,
southern Yukon. Canad Mineral 45: 203–227
Vianna P, da Costa G, Grave E, Evangelista H, Stern W
(2002a) Characterization of beryl (aquamarine variety) by
Mössbauer spectroscopy. Phys Chem Miner 29: 78–86
Vianna R, Jordt-Evangelista H, da Costa G, Stern W
(2002b) Characterization of beryl (aquamarine variety)
from pegmatites of Minas Gerais, Brazil. Phys Chem
Miner 29: 668–679
Wang RC, Che XD, Zhang WL, Zhang AC, Zhang H (2009)
Geochemical evolution and late re-equilibrium of Na-Csrich beryl from the Koktokay #3 pegmatite (Altai, NW
China). Eur J Mineral 21: 795–809
Whalen JB (1993) Geology, petrography, and geochemistry
of Appalachian granites in New Brunswick and Gaspésie,
Quebec. Geol Surv Canad Bull 436: 1–124
Whalen JB, Jenner GA, Longstaffe FJ, Hegner E (1996)
Nature and evolution of the eastern margin of Iapetus:
geochemical and isotopic constraints from Siluro–
Devonian granitoid plutons in the New Brunswick Appalachians. Can J Earth Sci 33: 140–155
Yang XM, Lentz DR, Chi G, Kyser K (2004) Fluid–
mineral reaction in the Lake George Granodiorite, New
Brunswick, Canada: implications for Au–W–Mo–Sb
mineralization. Canad Mineral 42: 1443–1464
67